Chimera Linux Pushes Ahead For FreeBSD User-Space Atop Linux, Built Using LLVM

[Raka555]

Wow - this sounds like the worst of both worlds - linux kernel with FreeBSD userland. My main complaint with FreeBSD is the lack of applications.

The presentation seems a bit confusing; for example I don't understand 'Not suckless' paired with 'If it sucks, get rid of it'. I do wish people would stick with one meaning of a word. And what does this mean "But technical dept is". Is what!?l Or "Neither is bootstrapping" Huh?!

And Gnome is the root of all evil. I also don't have a problem with meritocracy, as long as everyone gets a voice - the most successful work environments I've been in were like that.

What I do like is the use of LLVM tool chain, Ickey talks about doing that with SerpentOS, and I prefer if to GCC. Mainly because GCC doesn't support objc2.0 or wasm, and I understand the error messages better in clang.

Suckless is actually a "thing" : https://suckless.org/

I suspect they are referring to that.

[sinepgib]

How does giving positions of power to the most competent people and keeping things on-topic hinder developing an operating system? History's kinda full of examples on what happens when you start giving out positions of power over non-meritocratic things like ethnicity, who you're related to, clan memberships and political leanings. Failed states and collapsed societies all around.

The thing is, the most competent people isn't necessarily the most technical.

EDIT: a major example is how much GUIs got worse in GNOME. There isn't a democracy there, but people who just pick what to do and whose authority comes from having worked there longer and contributed more. They listen for bugs, but not for design complains. That's why Mate eventually was done, after all.
I'll mention it again because before the edit it was only later in this post, but I'm not defending token hires (which GNOME admittedly kinda does), I'm just not equating not being a meritocracy, where you don't just have competent people in positions of power, but you _strictly_ correlate previous output with power as the only path to having any kind of voice, to going the token hire route. I'm defending a more democratic style, that's all. An outsider may have something valuable to say.

What cave have you been living in for the last 15 years or so? Linux has been perfectly usable by the wider public for that time

Oh really? I guess that's why it's the first choice for everyone.

systemd has it's issues but it's most certainly not duct-taped together

Precisely. And it's the main example of progress made by commercial interest rather than meritocratic neckbeards making the decisions. If it were for the meritocratic technical obsessed people we'd all be running devuan, which favors duct-taped crap.

and the main thing that's keeping it from usage by most people is a familiarity with windows and reverse chicken or the egg problem with user share and commercial software. To get that software you need users and to get users you need that software. Getting token hires to go about running the project like Caligula does absolutely nothing for it.

I'm not talking about token hires. But meritocracy is just at the other end of the spectrum. Meritocracy is all about the ones who did more getting to make all the choices, and that's what I find counterproductive. Of course I'm not promoting simply filling positions arbitrarily, but democracy may be a better fit for choosing the targets of a project. The technical parts should, obviously, remain technical, and design choices come from agreements based on technical arguments.
Regarding the egg and chicken, it's true to some extent, but it's not the only reason.
First, I know tons of programmers running Ubuntu at work because that's what they give them to develop and still choosing Windows for home use.
This means it's not just familiarity in their case, they're capable of running Linux for a big fraction of their time and still choose Windows whenever they get a choice.
Besides, the lack of software is not just caused by reduced market share (although it certainly is a big factor). There are technical and social reasons why it's harder to support Linux than it is to support Windows.
For a start, when it comes to drivers it's either open sourcing or give mediocre support due to unstable kernel ABIs.
When it comes to userspace, it's simply infeasible to test against the whole range of distros and versions, even if you restrict to the popular ones, and saying "we support Linux" will both flood you with mails from users of fringe distros (and not only from people who choose niche distros, but some vendors in my country actually used to ship some random distro called RxArt) who find bugs in their particular flavor and make you lose reputation because of that. So you'd rather not advertise Linux support and call it a day.
This proliferation of distros, as well as the choice of ignoring ABI stability requests, come mostly from technical people in meritocratic projects where external input is disregarded in virtue of being sent by people who doesn't directly work for the project.

Uuh... You do remember I just wrote that I find this pointless, but at least they won't be causing problems for actually useful projects. At least until this project inevitably implodes after they go at each others' throats over politics.

That's cool, because I was kinda agreeing with you in this part, not arguing. While I don't think strict meritocracy is good for a project to really thrive, it kinda is the fairest way to lead a hobby one. It's welcoming ("wanna play a bit with this? come hang out!") but not imposing ("you should do this and that for me, while I sit and watch").

[sinepgib]

Politics should stay away from science and technology.

I disagree completely. Let me put it this way, and see if it makes sense to you.

There's an economic reality: resources are scarce. Science and technology need resources to make any progress. Resources that are created by someone else's work, specially for the case of science that is most often backed by States rather than private businesses.

Because resources are scarce, and most resources come from activities other than what you want to make, you need to give something in exchange to incentivize that movement. I.e., you need to focus on solving someone's problems.

Now, how do you choose what to work on? Who gets to make that choice?
If only the skilled people gets to choose, who are much fewer than both the domain space of problems that need solving and than the number of people who aren't skilled enough or work on different things, you'll probably get something that satisfies only them.
If only the people with the needs gets to choose, you'll probably end up in a situation where both skilled people is insufficient (because of what I said in the previous line) and solutions are really infeasible (because the people skilled to solve the problem is also generally the people who knows which solutions are viable).

So, we need to come up with a way to prioritize problems to solve with limited resources. Because resources generally come from the work of the masses, how to use them becomes a political problem.

Once you picked a problem, where you absolutely need to convince people outside your project to invest resources in it, the how is up to the ones who do the work. That part can be as meritocratic as you want, but that only makes sense at the point of making the final call after technical solutions get discussed. Even at the technical level, with no discussion between the makers you're bound to make big mistakes.

With no politics, you have no science. But with too much politics, you have ineffective science. There's a need for balance, but there's also a need for democracy within projects and the ability to listen to external feedback.

[brad0]

I don't mean to encourage the devs, but, just another distro who's trying to be different.

Man, i don't know whether choices on linux distro is good or not. Most of them don't bring anything new to the table.

Because the community needs more of the thousands of the same ol' same ol'.

[brad0]

So its like GNU/Linux, with a worse userland and compiled by a worse compiler. Now that's a compelling package.

Good, it stays away from crappy GCC and the other crappy bits of "traditional" GNU/Linux.

[q66_]

What a bad name…

• https://chimeraos.org/
• https://chimera-linux.org/

nothing i can do about this one; the project had already been in public development for about a month when that ChimeraOS thing renamed (it was called GamerOS previously)

[Luke_Wolf]

*snip*

You are posting the most idiotic garbage I have ever seen on this forum, and that is saying a lot.

The very meritocracy you're shitting yourself over is the very same one that forced universal adoption of systemd, the very same one that will force universal adoption in the long run of Wayland unless something better comes along in the meantime. Meritocracy is the force that determined that while Devuan may exist there is no interest and no manpower supporting it. It is the meritocracy that decided that Devuan is a project that is only fit for making Michael money through clicks and forum posts, but even then the return on investment for inflammatory systemd articles has dropped off at this point.

It is meritocracy you have to thank for the modern Linux desktop being in as good of shape as it is, and if we look over at Debian democracy only slowed things down, it forced distros to adopt less ideal solutions than the solutions provided by the people who stepped up and did the work (Red Hat, Lennart, etc).

Your attempt to tie meritocracy to devuan is not only backwards but demonstrates both a keen lack of intelligence and a hard political agenda that due to the lack of intelligence you fail to hide in your incoherent ramblings.

[sinepgib]

You are posting the most idiotic garbage I have ever seen on this forum, and that is saying a lot.

So, do I get a prize or what?

[Luke_Wolf]

So, do I get a prize or what?

[L_A_G]

The thing is, the most competent people isn't necessarily the most technical.

We're talking about giving positions to the most qualified/competent people. If it's a project management job of course skills and experience in project management is going to be highly relevant to the job. I'm not going to call your argument a straw man because I'm not quite sure you even realize what you're talking about trying to bash the concept of hiring and promoting the most qualified people.

EDIT: a major example is how much GUIs got worse in GNOME. There isn't a democracy there, but people who just pick what to do and whose authority comes from having worked there longer and contributed more. They listen for bugs, but not for design complains. That's why Mate eventually was done, after all.

The thing about user input is that you need to figure out what to listen to. You can't listen to everything because any large project is going to get feedback that's highly contradictory at times. Design is also something that works very poorly when done by committee with a large number of people contributing and only really works when done by a relatively small group with a focused vision. This can obviously go awry, but unlike design by committee it's not guaranteed to do so.

I'll mention it again because before the edit it was only later in this post, but I'm not defending token hires (which GNOME admittedly kinda does), I'm just not equating not being a meritocracy, where you don't just have competent people in positions of power, but you _strictly_ correlate previous output with power as the only path to having any kind of voice, to going the token hire route. I'm defending a more democratic style, that's all. An outsider may have something valuable to say.

Now I'm just going to straight up call you out for using very weak straw men. Meritocracy does not mean you don't listen to user feedback and it most certainly isn't just about who's been involved in a project the longest. Claiming the former is like claiming that keeping the kitchen in a restaurant clean means you won't get the customer orders right.

Oh really? I guess that's why it's the first choice for everyone.

So now we're doing the argumentum ad populum fallacy. What's the most popular is the best. By that logic McDonalds is the world's best restaurant and we both know McDonalds isn't exactly Michelin star material...

Precisely. And it's the main example of progress made by commercial interest rather than meritocratic neckbeards making the decisions. If it were for the meritocratic technical obsessed people we'd all be running devuan, which favors duct-taped crap.

You do know that systemd is very much an open source thing developed very much like Gnome, which you yourself lampooned in the same comment as an example of a bad open source project. They most certainly don't listen the criticism of it being so badly riddled with functional dependencies that it's effectively monolithic in a way that pulling in one part results in you having to pull in damn near the whole thing and replace every part that it implements with its' implementation.

I'm not talking about token hires. But meritocracy is just at the other end of the spectrum. Meritocracy is all about the ones who did more getting to make all the choices, and that's what I find counterproductive. Of course I'm not promoting simply filling positions arbitrarily, but democracy may be a better fit for choosing the targets of a project. The technical parts should, obviously, remain technical, and design choices come from agreements based on technical arguments.

Again, meritocracy is about hiring the most competent and qualified people for a particular role. It doesn't preclude input from outside parties when making decisions nor does it limit relevant experience to just within that particular organization. The only person who would claim something like that would be someone who's angry they, as a total outsider, couldn't just come in and start making important decisions.

For a start, when it comes to drivers it's either open sourcing or give mediocre support due to unstable kernel ABIs.

Unstable kernel ABI's has not been an issue for a very very long time... Not sure where you've heard this, but whoever told you this doesn't know what they're talking about.

When it comes to userspace, it's simply infeasible to test against the whole range of distros and versions, even if you restrict to the popular ones, and saying "we support Linux" will both flood you with mails from users of fringe distros (and not only from people who choose niche distros, but some vendors in my country actually used to ship some random distro called RxArt) who find bugs in their particular flavor and make you lose reputation because of that. So you'd rather not advertise Linux support and call it a day.

This is literally a talking point from Microsoft's infamous "Halloween documents" FUD campaign. From the perspective of an application or application developer there really hasn't been much of a difference between distros outside of some really niche ones for, again, a very very long time. Whoever you've heard this from either doesn't know what they're talking about or is literally a Microsoft sales person.

This proliferation of distros, as well as the choice of ignoring ABI stability requests, come mostly from technical people in meritocratic projects where external input is disregarded in virtue of being sent by people who doesn't directly work for the project.

Seriously, am I talking to a time traveler from 20 years ago here? Because this repeating of Microsoft's anti open source FUD from the late 90s and early 2000s is something I haven't seen for a very long time. People know better in 2022.

That's cool, because I was kinda agreeing with you in this part, not arguing. While I don't think strict meritocracy is good for a project to really thrive, it kinda is the fairest way to lead a hobby one. It's welcoming ("wanna play a bit with this? come hang out!") but not imposing ("you should do this and that for me, while I sit and watch").

Again, the fact that you don't like that you're not put in charge of things right as you come in trough the door doesn't mean that the concept of hiring of the most competent and qualified is fatally flawed. Get over yourself, work on your resume by actually accomplishing things and eventually you'll get on top. But you're not going to get hired over people who've actually accomplished things when you've yet to do that.